Re: full-text indexing

> On Tue, 18 Apr 2000, Bruce Momjian wrote:
>
> > I have one word for you: CLUSTER. Without it, index lookups are too
> > slow. With it, they are rapid. I have done some work like this
> > commerically with Ingres, which has an ISAM type that keeps the matching
> > rows pretty close on a newly-created ISAM index. In PostgreSQL, and
> > regular CLUSTER will keep you good.
>
> I agree! The last bit of advice given in the full text README. As I
> said, I'd built full-text stuff for experimentation (I had maybe 30k of
> raw text, which amounted to several 100,000 indexed entries), and I had
> clustered it, and it was pretty darn fast, even on a Pentium 233 with only
> 48 megs of RAM. I have significantly better hardware to run it on now.
> The original project called MySQL, but it just didn't have what we needed
> to put something like this together.

With the original author, testing was fast, but when he loaded all the
data, it got very slow. The problem was that as soon as his data
exceeded the buffer cache, performance became terrible.

--
Bruce Momjian | http://www.op.net/~candle
pgman(at)candle(dot)pha(dot)pa(dot)us | (610) 853-3000
+ If your life is a hard drive, | 830 Blythe Avenue
+ Christ can be your backup. | Drexel Hill, Pennsylvania 19026